Divalent Cation Removal by Donnan Dialysis for Improved ReverseElectrodialysisTimon Rijnaarts,†,‡ Nathnael T. Shenkute,† Jeffery A. Wood,§ Wiebe M. de Vos,†

and Kitty Nijmeijer*,∥

†Membrane Science & Technology, University of Twente, MESA+ Institute for Nanotechnology, Drienerlolaan 5, P.O. Box 217, 7500AE Enschede, The Netherlands‡Wetsus, European Centre of Excellence for Sustainable Water Technology, Oostergoweg 9, 8911 MA Leeuwarden, The Netherlands§Soft Matter, Fluidics and Interfaces, University of Twente, MESA+ Institute for Nanotechnology, Drienerlolaan 5, P.O. Box 217,7500 AE Enschede, The Netherlands∥Membrane Materials and Processes, Department of Chemical Engineering and Chemistry, Eindhoven University of Technology, HetKranenveld 14, P.O. Box 513, 5600 MB Eindhoven, The Netherlands

*S Supporting Information

ABSTRACT: Divalent cations in feedwater can causesignificant decreases in efficiencies for membrane processes,such as reverse electrodialysis (RED). In RED, power isharvested from the mixing of river and seawater, and theobtainable voltage is reduced and the resistance is increased ifdivalent cations are present. The power density of the REDprocess can be improved by removing divalent cations fromthe fresh water. Here, we study divalent cation removal fromfresh water using seawater as draw solution in a Donnandialysis (DD) process. In this way, a membrane system withneither chemicals nor electrodes but only natural salinitygradients can be used to exchange divalent cations. For DD, the permselectivity of the cation exchange membrane is found to becrucial as it determines the ability to block salt leakage (also referred to as co-ion transport). Operating DD using a membranestack achieved a 76% reduction in the divalent cation content in natural fresh water with residence times of just a few seconds.DD pretreated fresh water was then used in a RED process, which showed improved gross and net power densities of 9.0 and6.3%, respectively. This improvement is caused by a lower fresh water resistance (at similar open circuit voltages), due toexchange of divalent for monovalent cations.

KEYWORDS: Calcium, Magnesium, Cation exchange membrane, Donnan dialysis, Reverse electrodialysis, Hardness removal

■ INTRODUCTION

In the treatment of water there is an increasing interest inpretreatment technologies to enhance process efficiencies. Forreverse electrodialysis (RED), where the controlled mixing offresh water and seawater is used to harvest power, divalentcations in river water can significantly decrease RED perform-ance.1−3 In membrane-based processes that use natural waters,scaling can lead to lower efficiencies. Typical scaling consists ofthe divalent cations Ca2+ and Mg2+ with the anions SO4

2− andCO3

2−, which form salts with low solubilities leading toprecipitation.4−7 These precipitates cause membrane fouling,which decreases water recoveries for filtration or currentefficiencies for ED processes.It is therefore of interest to remove divalent cations prior to

these hardness-sensitive processes. Divalent cations can beprecipitated by chemical precipitation or exchanged byprocesses such as ion exchange (IEX) or Donnan dialysis(DD).8 In a typical IEX process, a column loaded with ion-exchange resin is preconditioned with monovalent ions such as

Na+ or H+, which can exchange for undesired divalent cations.The total charge of the ions in solution (concentrationmultiplied by valence of the ions) does not change; however,divalent cations are exchanged for monovalent cations in theresin. The driving force of this exchange is entropic gain:release of two monovalent cations by capturing a single divalentcation. After using IEX to exchange undesired divalent cations,the IEX resin has to be regenerated with a highly concentratedmonovalent cation (or acid) containing stream. Typically, twoIEX columns are required so one can exchange divalent cationswhile the other is regenerated. This doubles capital expensesand in addition requires chemicals (brine or acid) to regeneratethe IEX resin.8

In contrast, DD is a technique that can exchange divalentcations continuously without the need for regeneration.9 The

Received: February 23, 2018Revised: March 25, 2018Published: April 5, 2018

Research Article

pubs.acs.org/journal/ascecgCite This: ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

© 2018 American Chemical Society 7035 DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

This is an open access article published under a Creative Commons Non-Commercial NoDerivative Works (CC-BY-NC-ND) Attribution License, which permits copying andredistribution of the article, and creation of adaptations, all for non-commercial purposes.

Dow

nloa

ded

via

80.6

5.12

2.18

6 on

Jul

y 6,

202

0 at

13:

50:0

9 (U

TC

).Se

e ht

tps:

//pub

s.ac

s.or

g/sh

arin

ggui

delin

es f

or o

ptio

ns o

n ho

w to

legi

timat

ely

shar

e pu

blis

hed

artic

les.

same mechanism is found in biological cell membranes, whereNa+/Ca2+ exchangers use a gradient of Na+ to transport Ca2+

against its concentration gradient.10 This exchange processrelies on ion exchange membranes, instead of ion exchangeresins as used for IEX. The solution with undesired divalentcations is brought into contact with a cation exchangemembrane (CEM), and on the other side of the CEM, thereis a draw solution containing a high concentration ofmonovalent cations. This configuration allows for exchange ofthe divalent cations with monovalent cations in the drawsolution. The driving force behind this process is a similarentropic gain as found in IEX, resulting in high Donnanpotentials for monovalent species.11−14

Recent studies on divalent cation exchange pretreatmentshave been performed by Vanoppen et al. for reverse-osmosis(RO), using IEX as well as DD.8,15 In their work they foundIEX to be substantially more cost-effective due to the high priceof ion exchange membranes (300 €/m2) required for the DDprocess. However, they calculated that if the price of ionexchange membranes decreases to ∼10 €/m2, the processbecomes economically interesting, which is a price rangecomparable to RED.16

Previous work on RED has shown that divalent cations are amajor challenge in this process and limit the obtainable powerdensities due to both membrane resistance increase and uphilltransport.1,2 The use of new membranes that do not suffer fromsignificant membrane resistance increase overcomes the firstlimitation.17 However, uphill transport remains a problem dueto the presence of divalent cations in the river water. Uphilltransport is the exchange of divalent cations in the river wateragainst their concentration gradient with monovalent cations inthe seawater, which effectively lowers the obtainable voltage.Monovalent-selective CEMs could counter uphill transport;however, recent work showed that these monovalent selectiveCEMs have an increase in membrane resistance over time whilemultivalent permeable CEMs have stable resistances for longertimes.18 Besides uphill transport, the low conductivity of theriver water also limits power output. For a typical RED stack,the dominant resistance is that of the river water.19−22 Thereare two reasons for this low conductivity: First, the ionicconductivity of the river water is low, and therefore thiscompartment resistance in the RED stack is high. Recentstudies have focused on decreasing this resistance either usingdecreased spacer thickness19 or by introducing dynamic spacerto lower the river water compartment thickness.22 As such, itwas possible to reduce the resistance of this compartment, as itsresistance scales linearly with compartment thickness. Second,the low river water concentration combined with a largeconcentration gradient between the seawater and the riverwater on either side of the membrane induces an increasedmembrane resistance in both AEMs and CEMs.20,21 Both ofthese effects stress the importance of decreasing the river waterresistance to decrease the overall stack resistance. DD cantherefore be an attractive process to alleviate these resistancesby exchanging divalent cations for monovalent cations with ahigher activity (one divalent cation is exchanged for twomonovalent cations), which decreases resistance.In this work, we study the applicability of DD to exchange

divalent cations from river water by using seawater as a drawsolution, as shown in Figure 1. Studies using artificial naturalwaters are performed with different cation exchange mem-branes to show the proof-of-principle and to investigate theinfluence of membrane properties on DD performance. The

effect of DD pretreatment on RED power harvesting is thenassessed with natural river and seawater and compared totheoretical models to determine the maximum possibleenhancement for RED with DD pretreatment.

■ THEORYIn this section, a detailed background of DD is given tounderstand the principles and to predict divalent cationexchange. The influence of river water ion concentration onRED performance is predicted to determine the theoreticalbenefits of using DD pretreatment as well.

DD. The driving force of DD is an entropic gain byexchanging a single divalent cation for two monovalent cationsat the low concentration side. This entropic gain can beanalyzed by Donnan potentials of the different ionic species.For each ionic species “i”, Donnan potentials can be calculatedby their respective concentrations on each side of themembrane (eq 1).8,9,13 Assumptions in this equation are thatthere is negligible convection through the membrane, thatelectroneutrality is conserved, and that the membrane isperfectly permselective (no co-ion transport through themembrane). The system will strive toward equilibrium, drivenby entropy, and as such the Donnan potentials of each ionicspecies will equilibrate. This does not imply that theirconcentrations are equal; only the ratio of the ionic speciesto the power one over their respective valence will equilibrate.In the case of monovalent sodium (Na+, z = 1) and a divalentcation (M2+, z = 2) the following equilibrium will establish (eq2):

=⎛⎝⎜⎜

⎞⎠⎟⎟E

RTF

a

aln i

i

z

Don,c

,d

1/ i

(1)

=+

+

+

+

⎛⎝⎜⎜

⎞⎠⎟⎟

⎛⎝⎜⎜

⎞⎠⎟⎟

a

a

a

aNa ,c

Na ,d

1M ,c

M ,d

1/22

2 (2)

where EDon is the Donnan potential (V), R is the gas constant(8.314 J/(K·mol)), T is the temperature (K), F is the Faradayconstant (96 485 C/mol), z is the valence of the ion (−), and ais the activity of ion i (M) in the concentrate (c) or diluate (d)compartment.Donnan potentials depend on the respective activities of each

ion across the membrane (eq 2). Therefore, when the ionactivities in both streams are known, one can predict the

Figure 1. Schematic concept of DD as pretreatment, to exchangedivalent cations from river water and subsequent use in combinationwith RED. In the DD pretreatment step, exchange of divalent cationsfor monovalent Na+ from seawater is achieved.

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7036

direction of transport for each ion. For typical DD processes,the activity in the concentrate should be (at least) 10 timeslarger compared to that in the diluate to ensure sufficientdriving force.13 For this application, when using (artificial) riverand seawater, the activity ratio of ions in seawater and riverwater is ∼90. Donnan equilibria and concentrations over timeare given in Figure 2 for Na+ and Mg2+. For each mole of

divalent Mg2+ exchanged, double the quantity of moles of Na+

is exchanged until equilibrium is reached. This equilibrium canbe determined from Donnan potentials for Na+ and Mg2+ (seeFigure 2c) during the exchange. Initially, the Donnan potentialof Mg2+ is lower than Na+. This means that Na+ will move withits concentration gradient (from seawater to river water),whereas Mg2+ will move against its concentration gradient(from river water to seawater), to maintain electroneutrality.This exchange will continue until the Donnan potentials ofboth ions are equal.RED Model. DD treated water can subsequently be used as

input for RED to harvest power from the salinity gradient. ForRED, there is a trade-off between open circuit voltage (OCV),which increases at a lower river water salt concentration, andthe resistance (Rstack), which decreases at a higher river watersalt concentration. To predict the maximum obtainable powerdensity in RED after complete exchange of divalent formonovalent cations in the river water using DD, a previouslydeveloped model by Veerman et al. is adapted.23 The aim ofthis model is to understand the influence of the concentration

of salt in the river water. Specific details on the model are givenin the Supporting Information (SI 1).The dependences of the open circuit voltage (OCV), stack

resistance (Rstack), and power density (PD) are calculated as afunction of the river water NaCl concentration. In the model,the effect of divalent cations is excluded, as the aim of themodel is to establish the effect of complete divalent cationexchange by DD for RED. For natural river water, the NaClconcentration is 0.0035 M, which is well below the NaClconcentration with highest predicted power densities (between0.01−0.02 M for this system). This trade-off between opencircuit voltage and resistance in the stack is shown in Figure 3.

The maximum power density at a concentration of 0.014 MNaCl enables a 13% increase in gross power density comparedto natural river water (0.0035 M NaCl). In the Results andDiscussion section, the experimental results obtained with DDpretreated water will be compared with this predictive model.

■ MATERIALS AND METHODSMembranes and Chemicals. CEMs, Neosepta standard-grade

CMX (Eurodia, France), multivalent-permeable T1, and standard-grade Type II CEMs (FUJIFILM, The Netherlands) were used for DDexperiments; only T1 membranes are used for RED experiments.Anion exchange membranes (AEMs) Type I were used for the REDexperiments (FUJFILM, The Netherlands). Potassium chloride,magnesium chloride hexahydrate, anhydrous calcium chloride, andpotassium hexacyanoferrate (III) and potassium hexacyanoferrate (II)trihydrate, which act as a redox couple for RED, were purchased fromSigma-Aldrich. Sodium chloride (pharmaceutical grade) was kindlysupplied by AkzoNobel (Hengelo, The Netherlands).

DD in Diffusion Cells and Stacks. For lab-scale DD, diffusioncells with compartment volumes of 65 mL (with magnetic stirrers) andan active membrane area of 11.3 cm2 were used. Three different CEMswere evaluated. These CEMs were soaked in the same draw solutionsas used for DD for 24 h prior to use. Samples of river and seawatercompartments were taken over time to monitor the cationconcentrations. To scale-up and operate DD continuously, a crossflowRED stack (REDstack BV, The Netherlands) with a 10 × 10 cm2

active area was used. Both setups are illustrated in Figure 4. Four CMXmembranes were used to create two channels to feed fresh water andthree channels to feed seawater. In this manner both CEMs in contactwith the fresh water channel were used for ion exchange.

Figure 2. Theoretical batch DD process of river water with seawater asdraw solution. Top figures show the concentrations over time in theriver water (a) and seawater (b), respectively, where initialconcentrations are for natural waters. Part c shows Donnan potentialsof Na+ and Mg2+ over time. The large potential of Na+ (122 mV)compared to Mg2+ (57 mV) drives the exchange of Mg2+ uphill.Finally, at equilibrium, the Donnan potential of each cation is equal (inthis case around 115 mV). Part d shows the initial cationconcentration profiles in the seawater, CEM, and river water,respectively, calculated from the initial Donnan equilibrium. Straightarrows indicate transport direction for each cation and dashed arrowsshow equilibrium.

Figure 3. Modeled trade-off between RED stack performance andconcentration of NaCl in river water in M. The graph shows theoptimum between open circuit voltage (OCV) and cell area resistance(R). The optimum for gross power density (PD) is between 0.01 and0.02 M with a maximum at 0.014 M. Seawater concentration (notshown) is at natural concentrations for the Waddensea (NL) (0.40M).

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7037

Polyamide woven spacers, to separate the membranes in DD andRED (Deukum GmbH, Germany), had a thickness of 260 μm and afree void fraction (volume) and free surface fraction (projectedsurface) of 0.726 and 0.476, respectively.Ion chromatography (IC) was used to analyze the cation

concentrations (Metrosep C6-150/4.0 column in a Metrohm 850Professional IC with eluent of 1.7 mM HNO3 and dipicolinic acid).Cation samples were diluted with 2 mM HNO3 prior to analysis.RED Stack Performance Evaluation. Reverse electrodialysis

experiments were performed with 10 cell pairs of T1 CEM and Type IAEMs. These membranes were soaked for 24 h in 0.5 M NaCl prior toassembly in the stack. The T1 CEM was selected because of itssuperior RED performance in waters with divalent cations; however, itstill suffers from uphill transport.17

A cross-flow RED stack (REDstack BV, The Netherlands) with anactive area of 6.5 × 6.5 cm2 was equipped with Pt-coated Ti meshelectrodes (Magneto Special Anodes BV, The Netherlands). Thewater composition obtained from DD was used as feed in the REDstack. The temperature was controlled (20 ± 1 °C) by a Julabo F12-ED thermostat, and pulsation dampeners (in-house built) were used tosuppress pulsations caused by the peristaltic feedwater pumps (ColeParmer). An in-house built flow meter (McMillan Co. 101 flo-sen) wasused to measure the flow rate. For the electrode rinse, a solution of0.25 M NaCl with 0.1 M K4Fe(CN)6 and 0.1 M K3Fe(CN)6 was used.An overpressure of 0.1 bar on the electrolyte was applied to ensuremembrane and spacer packing.Electrochemical analysis of the stack was performed as described in

previous work.17 First, membranes were equilibrated with the feedsolution for 20 min under current (20 A/m2). Next, the OCV andresistance by alternating current (AC) were measured. For the ACresistance, three measurements were performed at 10, 5, and 1 kHz inthat order, at an amplitude of 0.01 A (2.4 A/m2) to measure the ohmicresistance of the stack. Finally, for the direct current (DC) resistance(including nonohmic resistance) and the power densities, 10 currentsteps from 0 to 50 A/m2 and 10 current steps back to 0 A/m2 wereapplied to measure the total resistance and calculate the stack powerdensities. Each current density was applied for 30 s before the voltagewas measured to allow for buildup of boundary layers. Membraneswere characterized on membrane resistance in 0.5 M NaCl using DCand permselectivity in 0.1/0.5 M NaCl.17

■ RESULTS AND DISCUSSION

In this study, we investigate the exchange of divalent cations byDD, as a pretreatment for RED. First, results on DD arereported using lab-scale diffusion cells with three differentCEMs. Based on the lab-scale study, the best membrane isselected for continuous operation in a membrane stack. Thescale-up of DD in a membrane stack is performed, followed byusing the pretreated water in a RED process to assess its effect.DD in Lab-Scale Diffusion Cells. Figure 5 shows the

cation concentrations in the river water during DD. Thedivalent cation concentrations decrease over time while the Na+

concentrations increase. After 180 min, the majority of divalentcations is removed (above the experimental detection limit of

0.1 mM). Correspondingly, in the seawater a small decrease inNa+ concentration and a small increase in divalent cationconcentrations is observed. Since the absolute concentrations inthe seawater are much higher than those in the river water, therelative change in these concentrations is small and thereforethe composition of the seawater is hardly changed. Theseresults demonstrate that the exchange of divalent formonovalent cations indeed occurs. Moreover, the reachedequilibrium concentrations are equivalent to those predicted bythe Donnan potential calculations, shown in the SupportingInformation (SI 2).These measurements are performed for three different

CEMs, and their performances are compared in Figure 6.The exchange rate should be as high as possible to allow forshort residence times. It is expected that the membraneresistance is a suitable predictor for exchange rates, as it is ameasure for ionic transport rate at a certain driving voltage.These driven voltages are defined by the concentrations in thesolutions and are therefore equal for all CEMs. However, thereis no clear correlation between the exchange rate and theelectrical resistance. This difference, we believe, is caused by thesignificant leakage for the T1 membrane, which happenssimultaneously with the exchange (see Figure 6a). The leakageis defined here as the total equivalent concentration (ionconcentration times charge) in the river water at time tnormalized to the starting concentration. Therefore, its effectiveexchange rates are lower than would be expected from its lowresistance. For the best CEM, CMX, 98% divalent cationremoval after 3 h with 40 mmol/(m2·h) is achieved.Next to the exchange, there is the (unwanted) leakage of salts

(due to co-ion (anion) diffusion) from the seawater throughthe CEM to the river water. The purpose of DD in this case isto exchange the divalent cations from the river water without amajor increase of the total salt content in the river water, as thiscould lower the RED power density due to a lower drivingforce (salinity gradient). The model in the Theory sectionpredicted an optimum NaCl concentration between 0.01−0.02M, and in Figure 5, DD reaches the lower limit of this range(0.010 M). The normalized leakage depends on permselectiv-ity; for 100% permselective membranes the normalized leakageshould remain close to 1 (the initial concentration). In Figure6, it is shown that the lowest permselective membrane T1, witha permselectivity of 90%, leaks more salt over time compared toType II and CMX with permselectivities of 98% and 99%,respectively, as was expected. It appears that there is a trade-offfor the divalent cation exchange rate between permselectivityand resistances, and a high (98%) permselectivity is desired toprevent significant leakage. For these reasons (high permse-lectivity and low leakage), the CMX membrane was selected forscale-up.

Figure 4. Diffusion cell (a) and stack (b) configurations for DD used in this study. Cation exchange membranes are shown in dark green, river wateris shown in dark blue, and seawater is shown in light blue; spacers and stir bars are shown in gray.

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7038

Scale-up of DD in Stacks. These lab-scale experimentsshow the potential of DD using cation exchange membraneswith seawater as draw solution to exchange divalent cationsfrom river water. Residence times or surface-to-volume ratios(∼17 m2/m3) are limiting factors in these cells, so to decreaseresidence times from hours to seconds, one must improve themembrane surface to treated volume ratio to obtain viableresidence times. These high surface-to-volume ratios arerequired for a continuous and high-throughput process withoutthe need for large vessels. For this reason, membrane stacks areinvestigated to improve surface-to-volume ratios compared withdiffusion cells. Crossflow RED stacks have a very high surface-to-volume ratio (∼7700 m2/m3 in this study), as the active areais 10 × 10 cm2 with a channel thickness of only 260 μm.Scale-up of DD was performed using CMX membranes, as

this membrane showed the lowest leakage and the highestexchange rate in diffusion cells. In Figure 7, the results for DDwith CMX in a stack are shown. At residence times of onlyseveral seconds, there is already a clear decrease in Ca2+ in theriver water through exchange for Na+, indicating theapplicability of larger-scale operation of these systems. Similardivalent cation exchange rates are observed as in the diffusioncells (44 mmol/(m2·h) at 11.3 s). At residence times largerthan 7 s, the exchange appears to level off. Initially, the diluate

compartment consists of nearly 40 mol % of divalent cations;however, after DD the divalent cation content is reduced tobetween 13 to 4 mol % (divalent cation removal of 45 to 76%)at an exchange rate of 100−40 mmol/(m2·h) depending on

Figure 5. Concentrations of cations in the DD process with natural river (left) and seawater (right) across CMX in a small scale diffusion cell overtime. Lines are shown to guide the eye. Measurements are triplicates and errors are typically below 1%.

Figure 6. Schematic drawing of the exchange and leakage processes occurring in DD (a). Divalent cation exchange rates (b) and normalized leakage(c) of CMX, Type II, and T1 CEMs are shown as a function of log time. Resistances shown in the exchange rate graph are membrane arearesistances (b) and percentages shown in the leakage graph are the permselectivities of each membrane (c). In the graph with exchange rateslogarithmic trend lines and in the figure with leakage linear trend lines are plotted for visual aid.

Figure 7. Concentrations as a function of residence time for the scale-up DD in the river water compartment. Short (DD-S) and long (DD-L) residence times for DD are indicated by gray bars; both are used forRED pretreatment. Relative errors of the cation concentrations arebelow 1%, and for clarity error bars are omitted.

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7039

residence time (from 2.7 to 11.3 s), shown in SupportingInformation (SI 2). Similar removal results have been obtainedin recent work on DD by Vanoppen et al., who achieveddivalent cation removal of 70−80% at a rate of 20 mmol/(m2·h) using Fuji Type II CEMs in a stack.8

The results shown here demonstrate that DD is scalable andthat divalent cations can be removed with residence times ofjust a few seconds. Detailed comparison between DD in the celland in the stack is given in the Supporting Information (SI 2).RED Stack Performance. Finally, the effect of using DD

pretreated river water for RED is investigated. Initially, REDexperiments with benchmark concentrations (0.5 M vs 0.017M) of NaCl are done for comparison with the modelcalculations and literature.2,17 The model river and seawaters,as well as treated water by DD at a short (DD-S: 2.7 s) andlong DD residence time (DD-L: 11.3 s), are used to evaluatethe obtainable energy through RED. Water compositions for allstreams are provided in the Supporting Information (SI 3).In Figure 8, the measured OCV and cell area resistances are

plotted together with the calculated values from the model. The

two extremes of DD are shown in the graph. Longer DDresidence times allow for longer exchange times for divalentcations with Na+, resulting in lower divalent cation concen-trations. Effectively, the OCV in RED remains very similarbefore and after DD. This was unexpected as the DDpretreatment decreases divalent cation concentrations andthus should decrease the effect of uphill transport. Thisexpected increase in OCV is shown in the model predictions(see Figure 8). However, we also exchange these divalentcations for Na+, and therefore the concentration gradient inRED decreases as well. In other words, the overall result is thatthe mitigation of divalent cation uphill transport and thedecrease in Na+ gradient counterbalance each other and theOCV is minimally affected.However, the cell resistance is just as important for

determining the obtainable power density. In benchmark riverwater (0.017 M NaCl), this resistance is low due to therelatively high conductivity of the river water compartment. Incontrast, natural river water contains divalent cations and alower overall salt concentration, which result in a higherresistance. The river water treated by DD has a higherconcentration of Na+, which has higher activity compared to

divalent cations, and therefore the resistance of the river watercompartment decreases. The predicted resistances are not inperfect agreement with the measured values, since divalentcations in water and membranes as well as gradient-inducedresistances are not included in the predictions. These divalentcations decrease river water conductivity but increasemembrane resistance, so the overall effect is not straightforwardto predict. Qualitatively, the predictions do match the datashowing that more exchanged Na+ for divalent cationsdecreases the cell resistance.One could argue that simply increasing the river water

conductivity has the same effect. To test this, seawater is mixedwith river water to a conductivity equal to the one reached byDD-L, but with different composition (high divalent cationconcentrations). RED stack resistances for this water are closeto DD treated river water, but the OCV drops significantly dueto both a high divalent cation concentration (uphill transport)and a lower overall gradient. This causes a decrease in powerdensity of 21% compared to untreated water (SupportingInformation (SI 4)).In the end, power density is the main performance parameter

for RED as it takes both resistance and OCV effects intoaccount. The OCV hardly changes while resistance decreasesupon DD pretreatment. Gross power densitiesthe poweroutput per m2 of membrane cell pair areawith DD can beimproved by 1.4 to 9.0% compared to the case without DD asdetermined by experiments. By using a model, a 13%improvement in gross power density is predicted for DD-L at0.010 M NaCl. With the DD pretreatments in this study, not alldivalent cations are exchanged; hence, in the RED experimentsthere are still ∼25% divalent cations present, which explains thelower improvements as compared to the predictions. Theseimprovements in power density are achieved without anyaddition of chemicals (such as for IEX) or electrical energy(such as ED), by utilizing only a concentrated stream that isalready present. The cost of DD, however, is an extrapretreatment step, which does require pumping energy. InFigure 9, the net power density after subtracting pumping

losses for DD and RED is shown. At high flow velocities (DD-S) there is a large pressure drop and thus loss in net powerdensity, whereas at low flow velocities (DD-L) this pressuredrop is smaller. DD-L gives a net power density gain of 6.3%over the case without DD pretreatment.

Figure 8. Cell OCV and cell area resistance versus river water NaClconcentration. Data (◆ = OCV, ● = area resistance) and modelpredictions (lines) are shown. Natural is untreated river water andDD-S and DD-L denote short and long DD pretreatment. Measure-ments are performed in triplicate, and error bars are standarddeviations.

Figure 9. Net power densities with 100% NaCl, natural waters, andDD (short and long residence times) pretreated river water.Measurements are performed in triplicate, and error bars are standarddeviations.

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7040

This work aimed to provide a proof of principle of usingseawater to exchange divalent cations in the river water aspretreatment for RED. In this research, there is still room foroptimization of the DD stack. Higher surface-to-volume ratiosand enhanced mixing without drastic increases in pressuredrops would allow further improvements in removal of divalentcations in DD. This could be achieved by using mixingpromotors, such as profiled membranes.24 In that way, moredivalent cations can be exchanged, allowing the experimentalpower density improvements to approach the calculated 13%power density improvement. Furthermore, less scaling isanother benefit of this pretreatment.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acssusche-meng.8b00879.

Additional information on the RED model, Donnanpotentials, and surface contact and residence times; scaleup comparison between diffusion cells and stacks; andionic compositions for all streams as well as RED resultsfor premixing some seawater with river water (PDF)

■ AUTHOR INFORMATIONCorresponding Author*Kitty Nijmeijer. E-mail: d.c.nijmeijer@tue.nl.ORCIDJeffery A. Wood: 0000-0002-9438-1048Kitty Nijmeijer: 0000-0002-1431-2174NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was performed in the cooperation framework ofWetsus, European Centre of Excellence for Sustainable WaterTechnology (www.wetsus.eu). Wetsus is cofunded by theDutch Ministry of Economic Affairs and Ministry of Infra-structure and Environment, the Province of Fryslan, and theNorthern Netherlands Provinces. The authors would like tosincerely thank the participants of the research theme “BlueEnergy” for fruitful discussions and financial support. Theauthors would like to thank Dr. Sylwin Pawlowski fromUniversidade Nova de Lisboa for the many fruitful discussions.

■ REFERENCES(1) Post, J. W.; Hamelers, H. V. M; Buisman, C. J. N. Influence ofmultivalent ions on power production from mixing salt and fresh waterwith a reverse electrodialysis system. J. Membr. Sci. 2009, 330 (1−2),65−72.(2) Vermaas, D. A.; Veerman, J.; Saakes, M.; Nijmeijer, K. Influenceof multivalent ions on renewable energy generation in reverseelectrodialysis. Energy Environ. Sci. 2014, 7 (4), 1434−1445.(3) Avci, A. H.; Sarkar, P.; Tufa, R. A.; Messana, D.; Argurio, P.;Fontananova, E.; Di Profio, G.; Curcio, E. Effect of Mg2+ ions onenergy generation by Reverse Electrodialysis. J. Membr. Sci. 2016, 520,499−506.(4) Turek, M.; Dydo, P. Electrodialysis reversal of calcium sulphateand calcium carbonate supersaturated solution. Desalination 2003, 158(1), 91−94.(5) Bazinet, L.; Araya-Farias, M. Effect of calcium and carbonateconcentrations on cationic membrane fouling during electrodialysis. J.Colloid Interface Sci. 2005, 281 (1), 188−196.

(6) Casademont, C.; Pourcelly, G.; Bazinet, L. Effect of magnesium/calcium ratio in solutions subjected to electrodialysis: Characterizationof cation-exchange membrane fouling. J. Colloid Interface Sci. 2007, 315(2), 544−554.(7) Rozan ska, A.; Wisniewski, J. Modification of brackish watercomposition by means of Donnan dialysis as pretreatment beforedesalination. Desalination 2009, 240 (1), 326−332.(8) Vanoppen, M.; Stoffels, G.; Demuytere, C.; Bleyaert, W.;Verliefde, A. R. D. Increasing RO efficiency by chemical-free ion-exchange and Donnan dialysis: Principles and practical implications.Water Res. 2015, 80, 59−70.(9) Higa, M.; Tanioka, A.; Miyasaka, K. Simulation of the transport ofions against their concentration gradient across charged membranes. J.Membr. Sci. 1988, 37 (3), 251−266.(10) Liao, J.; Li, H.; Zeng, W.; Sauer, D. B.; Belmares, R.; Jiang, Y.Structural Insight into the Ion-Exchange Mechanism of the Sodium/Calcium Exchanger. Science 2012, 335 (6069), 686−690.(11) Miyoshi, H. Diffusion coefficients of ions through ion exchangemembrane in Donnan dialysis using ions of different valence. J. Membr.Sci. 1998, 141 (1), 101−110.(12) Nouri, S.; Dammak, L.; Bulvestre, G.; Auclair, B. Studies of thecrossed ionic fluxes through a cation-exchange membrane in the caseof Donnan dialysis. Desalination 2002, 148 (1), 383−388.(13) Velizarov, S.; Reis, M. A.; Crespo, J. G. Removal of trace mono-valent inorganic pollutants in an ion exchange membrane bioreactor:analysis of transport rate in a denitrification process. J. Membr. Sci.2003, 217 (1), 269−284.(14) Rozanska, A.; Wisniewski, J. Brackish water desalination withthe combination of Donnan dialysis and electrodialysis. Desalination2006, 200 (1), 615−617.(15) Vanoppen, M.; Stoffels, G.; Buffel, J.; De Gusseme, B.; Verliefde,A. R. D. A hybrid IEX-RO process with brine recycling for increasedRO recovery without chemical addition: A pilot-scale study.Desalination 2016, 394, 185−194.(16) Post, J. W.; Goeting, C. H.; Valk, J.; Goinga, S.; Veerman, J.;Hamelers, H. V. M; Hack, P. J. F. M. Towards implementation ofreverse electrodialysis for power generation from salinity gradients.Desalin. Water Treat. 2010, 16 (1−3), 182−193.(17) Rijnaarts, T.; Huerta, E.; van Baak, W.; Nijmeijer, K. Effect ofdivalent cations on RED performance and cation exchange membraneselection to enhance power densities. Environ. Sci. Technol. 2017, 51,13028.(18) Moreno, J.; Díez, V.; Saakes, M.; Nijmeijer, K. Mitigation of theeffects of multivalent ion transport in reverse electrodialysis. J. Membr.Sci. 2018, 550, 155−162.(19) Vermaas, D. A.; Saakes, M.; Nijmeijer, K. Doubled PowerDensity from Salinity Gradients at Reduced Intermembrane Distance.Environ. Sci. Technol. 2011, 45 (16), 7089−7095.(20) Galama, A. H.; Vermaas, D. A.; Veerman, J.; Saakes, M.;Rijnaarts, H. H. M.; Post, J. W.; Nijmeijer, K. Membrane resistance:The effect of salinity gradients over a cation exchange membrane. J.Membr. Sci. 2014, 467, 279−291.(21) Geise, G. M.; Curtis, A. J.; Hatzell, M. C.; Hickner, M. A.;Logan, B. E. Salt Concentration Differences Alter MembraneResistance in Reverse Electrodialysis Stacks. Environ. Sci. Technol.Lett. 2014, 1 (1), 36−39.(22) Moreno, J.; Slouwerhof, E.; Vermaas, D. A.; Saakes, M.;Nijmeijer, K. The Breathing Cell: Cyclic Intermembrane DistanceVariation in Reverse Electrodialysis. Environ. Sci. Technol. 2016, 50(20), 11386−11393.(23) Veerman, J.; Saakes, M.; Metz, S. J.; Harmsen, G. J. Reverseelectrodialysis: A validated process model for design and optimization.Chem. Eng. J. 2011, 166 (1), 256−268.(24) Pawlowski, S.; Rijnaarts, T.; Saakes, M.; Nijmeijer, K.; Crespo, J.G.; Velizarov, S. Improved fluid mixing and power density in reverseelectrodialysis stacks with chevron-profiled membranes. J. Membr. Sci.2017, 531, 111−121.

ACS Sustainable Chemistry & Engineering Research Article

DOI: 10.1021/acssuschemeng.8b00879ACS Sustainable Chem. Eng. 2018, 6, 7035−7041

7041